Skip to content

Open Whisperer 1.4.0 — Streaming TTS

Choose a tag to compare

View all tags
@PerIPan PerIPan released this 14 Jun 12:14
v1.4.0
4cb8a22

Open Whisperer 1.4.0

✨ New

  • Streaming text-to-speech — speech now starts after the first sentence is synthesized instead of waiting for the whole response, so the AI starts talking back noticeably sooner. Gapless playback, near-instant "hold on" barge-in, and automatic fallback to the classic method if streaming is unavailable.
  • Codex TTS respects your volume — the Codex CLI hook now honors the in-app volume setting instead of a fixed level.

🛠 Compatibility

  • Setup works in more environments — Python from pyenv, conda, MacPorts, or Intel Homebrew is now found correctly (setup previously could fail with "no Python found").

🐛 Stability & security

  • Fixed a transcription-overlay teardown crash (DispatchSource race).
  • Fixed mode-switch edge cases that could mis-route dictated text or leave the recorder stuck.
  • Fixed dictated text/focus landing in the wrong app when a menubar menu (ours or another utility's) was open at capture time.
  • Faster, cleaner app quit; clearer audio-engine error messages.
  • TTS failures are now logged; corrupt/partial audio is never played.
  • Tighter permissions on config files and removed an AppleScript-injection vector.

Install

Download OpenWhisperer-1.4.0.dmg below and drag it to Applications.

Bypass Gatekeeper (unsigned, ad-hoc build):

xattr -cr /Applications/Open\ Whisperer.app

Everything runs locally on Apple Silicon — no cloud, no data leaves your Mac.

Assets 3
Loading